The rock mass along the Jinsha River is relatively broken under complex geological action. Many ancient landslides were distributed along the Jinsha River in Gongjue County, which is very dangerous under the action of gravity, tectonic stress and river erosion. Efficient and accurate identification and monitoring of landslides is important for disaster monitoring and early warning. Interferometric synthetic aperture radar (InSAR) technology has been proved to be an effective technology for landslide hazard identification and mapping. However, great uncertainty inevitably exists due to the single deformation observation method, resulting in wrong judgment during the process of landslide detection. Therefore, to address the uncertainties arising from single observations, a cross-comparison method is put forward using SBAS-InSAR (small baseline subset InSAR) and PS-InSAR (permanent scatterers InSAR) technology. Comparative analysis of the spatial complementarity of interference points and temporal deformation refined the deformation characteristics and verified the reliability of the InSAR results, aiding in the comprehensive identification and further mapping of landslides. Landslides along the Jinsha River in Gongjue County were studied in this paper. Firstly, 14 landslides with a total area of 20 km² were identified by using two time-series InSAR methods. Then, the deformation characteristics of these landslides were validated by UAV (unmanned aerial vehicle) images, multiresource remote sensing data and field investigation. Further, the precipitation data were introduced to analyze the temporal deformation characteristics of two large landslides. Lastly, the influence of fault activity on landslide formation is further discussed. Our results demonstrate that the cross-comparison of the time-series InSAR method can effectively verify the accuracy of landslide identification.

The development of landslide hazards is spatially scattered, temporally random, and poorly characterized. Given the advantages of the large spatial scale and high sensitivity of InSAR observations, InSAR is becoming one of the main techniques for active landslide identification. The difficult problem is how to quickly extract landslide information from extensive InSAR image data. Since the instance segmentation model (Mask R-CNN) in deep learning can provide highly robust target recognition, we select the landslide-prone eastern edge of the Tibetan Plateau as a specific test area. Introducing and optimizing this model achieves high-speed and accurate recognition of InSAR observations. First, the InSAR patch landslide instance segmentation dataset (SLD) is established by developing a common object in context (COCO) annotation format conversion code based on InSAR observations. The Mask R-CNN+++ is found by adding three functions of the ResNext module to increase the fineness of the network segmentation results and enhance the noise resistance of the model, the DCB (deformable convolutional block) to improve the feature extraction ability of the network for geometric morphological changes of landslide patches, and an attention mechanism to selectively enhance usefully and suppress features less valuable to the native Mask R-CNN network. The model achieves 92.94% accuracy on the test set, and the active landslide recognition speed based on this model under ordinary computer hardware conditions is 72.3 km²/s. The overall characteristics of the results of this study show that the optimized model effectively enhances the perceptibility of image morphological changes, thereby resulting in smoother recognition boundaries and further improvement of the generalization ability of segmentation detection. This result is expected to serve to identify and monitor active landslides in complex surface conditions on a large spatial scale. Moreover, active landslides of different geometric features, motion patterns, and intensities are expected to be further segmented.

At present, Synthetic Aperture Radar Interferometry (InSAR) has been an important technique for active landslides recognition in the geological survey field. However, the traditional interpretation method through human–computer interaction highly relies on expert experience, which is time-consuming and subjective. To solve the problem, this study designed an end-to-end semantic segmentation network, called deep residual shrinkage U-Net (DRs-UNet), to automatically extract potential active landslides in InSAR imagery. The proposed model was inspired by the structure of U-Net and adopted a residual shrinkage building unit (RSBU) as the feature extraction block in its encoder part. The method of this study has three main advantages: (1) The RSBU in the encoder part incorporated with soft thresholding can reduce the influence of noise from InSAR images. (2) The residual connection of the RSBU makes the training of the network easier and accelerates the convergency process. (3) The feature fusion of the corresponding layers between the encoder and decoder effectively improves the classification accuracy. Two widely used networks, U-Net and SegNet, were trained under the same experiment environment to compare with the proposed method. The experiment results in the test set show that our method achieved the best performance; specifically, the F1 score is 1.48% and 4.1% higher than U-Net and SegNet, which indicates a better balance between precision and recall. Additionally, our method has the best IoU score of over 90%. Furthermore, we applied our network to a test area located in Zhongxinrong County along Jinsha River where landslides are highly evolved. The quantitative evaluation results prove that our method is effective for the automatic recognition of potential active landslide hazards from InSAR imagery.

Reservoir impoundment significantly impacts the hydrogeological conditions of reservoir bank slopes, and bank slope deformation or destruction occurs frequently under cyclic impoundment conditions. Ground deformation prediction is crucial to the early warning system for slow-moving landslides. Deep learning methods have developed rapidly in recent years, but only a few studies are on combining deep learning and landslide warning. This paper proposes a slow-moving landslide displacement prediction method based on the Informer deep learning model. Firstly, the Sentinel-1 (S1) data are processed to obtain the cumulative displacement time-series image of the bank slope by the Small-BAseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) method. Then, combining data on rainfall, humidity, and horizontal and vertical distances of pixel points from the water table line, this study created a dataset with landslide displacement as the target feature. After that, this paper improves the Informer model to make it applicable to our dataset. This study chose the Dawanzi landslide in the Baihetan reservoir area, China, for validation. After training with 50-time series deformation data points, the model can predict the displacement results of 12-time series deformation data points using 12-time series multi-feature data, and compared with the monitoring values, its Mean Square Error (MSE) was 11.614. The results show that the multivariate dataset is better than the deformation univariate data in predicting the displacement in the large deformation zone of bank slopes, and our model has better complexity and prediction performance than other deep learning models. The prediction results show that among zones I–IV, where the Dawanzi Tunnel is located, significant deformation with the maximum deformation rate detected exceeding –100mm/year occurs in Zones I and III. In these two zones, the initiation of deformation relates to the drop in water level after water storage, with the deformation rate of Zone III exhibiting a stronger correlation with the change in water level. It is expected that deformation in Zone III will either remain slow or stop, while deformation in Zone I will continue at the same or a decreased rate. Our proposed method for slow-moving landslide displacement forecasting offers fast, intuitive, and economically feasible advantages. It can provide a feasible research idea for future deep learning and landslide warning research.

The occurrence of surface strata movement in underground coal mining leads to the generation of numerous ground fissures, which not only damage the ecological environment but also disrupt building facilities, lead to airflow and easily trigger coal spontaneous combustion, induce geological disasters, posing a serious threat to people’s lives, property, and mining production. Therefore, it is particularly important to quickly and accurately obtain the information of ground fissures and then study their distribution patterns and the law of spatial-temporal evolution. The traditional field investigation methods for identifying fissures have low efficiency. The rapid development of UAVs has brought an opportunity to address this issue. However, it also poses new questions, such as how to interpret numerous fissures and the distribution law of fissures with underground mining. Taking a mine in the Shenfu coalfield on the semi-desert aeolian sand surface as the research area, this paper studies the fissure recognition from UAV images by deep learning, fissure development law, as well as the mutual feed of surface condition corresponding to the under-ground mining progress. The results show that the DRs-UNet deep learning method can identify more than 85% of the fissures; however, due to the influence of seasonal vegetation changes and different fissure development stages, the continuity and integrity of fissure recognition methods need to be improved. Four fissure distribution patterns were found. In open-cut areas, arc-shaped fissures are frequently observed, displaying significant dimensions in terms of depth, length, and width. Within subsidence basins, central collapse areas exhibit fissures that form perpendicular to the direction of the working face. Along roadways, parallel or oblique fissures tend to develop at specific angles. In regions characterized by weak roof strata and depressed basins, abnormal reverse-“C”-shaped fissures emerge along the mining direction. The research results comprehensively demonstrate the process of automatically identifying ground fissures from UAV images as well as the spatial distribution patterns of fissures, which can provide technical support for the prediction of ground fissures, monitoring of geological hazards in mining areas, control of land environmental damage, and land ecological restoration. In the future, it is suggested that this method be applied to different mining areas and geotechnical contexts to enhance its applicability and effectiveness.

Underground coal mining induces surface subsidence, which in turn impacts the stability of slopes in mountainous regions. However, research that investigates the coupling relationship between surface subsidence in mountainous regions and the occurrence of multiple surface hazards is scarce. Taking a coal mine in southwestern China as a case study, a detailed catalog of the surface hazards in the study area was created based on multi-temporal satellite imagery interpretation and Unmanned aerial vehicle (UAV) surveys. Using interferometric synthetic aperture radar (InSAR) technology and the logistic subsidence prediction method, this study investigated the evolution of surface subsidence induced by underground mining activities and its impact on the triggering of multiple surface hazards. We found that the study area experienced various types of surface hazards, including subsidence, landslides, debris flows, sinkholes, and ground fissures, due to the effects of underground mining activities. The InSAR monitoring results showed that the maximum subsidence at the back edge of the slope terrace was 98.2 mm, with the most severe deformation occurring at the mid-slope of the mountain, where the maximum subsidence reached 139.8 mm. The surface subsidence process followed an S-shaped curve, comprising the stages of initial subsidence, accelerated subsidence, and residual subsidence. Additionally, the subsidence continued even after coal mining operations concluded. Predictions derived from the logistic model indicate that the duration of residual surface subsidence in the study area is approximately 1 to 2 years. This study aimed to provide a scientific foundation for elucidating the temporal and spatial variation patterns of subsidence induced by underground coal mining in mountainous regions and its impact on the formation of multiple surface hazards.